How to use the workflows.cloudify_system_workflows.snapshots.utils.run function in workflows

flags = ["--exclude-table={0}".format(t)
                     for t in exclude_tables]
            flags.extend(["--exclude-table-data={0}".format(t)
                          for t in exclude_tables])
        pg_dump_bin = os.path.join(self._bin_dir, 'pg_dump')
        command = [pg_dump_bin,
                   '-a',
                   '--host', self._host,
                   '--port', self._port,
                   '-U', self._username,
                   db_name,
                   '-f', destination_path]
        if table:
            command += ['--table', table]
        command.extend(flags)
        run_shell(command)

def _fix_snapshot_ssh_db(tenant, orig, replace):
    python_bin = '/opt/manager/env/bin/python'
    dir_path = os.path.dirname(os.path.realpath(__file__))
    script_path = os.path.join(dir_path, 'fix_snapshot_ssh_db.py')
    command = [python_bin, script_path, tenant, orig, replace]
    res = run(command)
    if res and hasattr(res, 'aggr_stdout'):
        ctx.logger.debug('Process result: \n{0}'
                         .format(res.aggr_stdout))

def dump_status_reporter_roles(self, tempdir):
        ctx.logger.debug('Dumping status reporter users\' roles')
        path = os.path.join(tempdir, 'status_reporter_roles.dump')
        command = self.get_psql_command(self._db_name)

        query = (
            'select array_to_json(array_agg(row)) from ('
            'select * from users_roles where user_id in ({0})'
            ') row;'.format(_STATUS_REPORTERS_IDS_QUERY_TUPLE)
        )
        command.extend([
            '-c', query,
            '-t',  # Dump just the data, without extra headers, etc
            '-o', path,
        ])
        run_shell(command)
        ctx.logger.debug('Dumped status reporter roles to {}'.format(path))
        return path

def _restore_dump(self, dump_file, db_name, table=None):
        """Execute `psql` to restore an SQL dump into the DB
        """
        ctx.logger.debug('Restoring db dump file: {0}'.format(dump_file))
        command = self.get_psql_command(db_name)
        command.extend([
            '-v', 'ON_ERROR_STOP=1',
            '--single-transaction',
            '-f', dump_file
        ])
        if table:
            command += ['--table', table]
        run_shell(command)

def _prepend_dump(dump_file, queries):
        queries_str = '\n'.join(queries)
        ctx.logger.debug('Adding to beginning of dump: {0}'
                         .format(queries_str))
        pre_dump_file = '{0}.pre'.format(dump_file)
        new_dump_file = '{0}.new'.format(dump_file)
        with open(pre_dump_file, 'a') as f:
            f.write('\n{0}\n'.format(queries_str))
        # using cat command and output redirection
        # to avoid reading file content into memory (for big dumps)
        cat_content = 'cat {0} {1}'.format(pre_dump_file, dump_file)
        run_shell(command=cat_content, redirect_output_path=new_dump_file)
        return new_dump_file

def create_db(self):
        ctx.logger.debug('Creating db')
        create_db_bin = os.path.join(self._bin_dir, 'createdb')
        command = [create_db_bin,
                   '--host', self._host,
                   '--port', self._port,
                   '-U', self._username,
                   '-T', 'template0',
                   self._db_name]
        run_shell(command)